The research plans for next year involve three projects. The problem of what makes a text hard or easy to read is as yet very poorly understood. We know some factors (number of propositions, number of concepts), but it would be important to continue this type of work, with the eventual goal of deriving readability formula for texts that is theoretically motivated. The second topic concerns different levels of memory for texts and will also be explored further in the next year. The problem is important not only methodologically, but also because it provides a direct link between list-learning studies of memory and verbal learning and studies using textual materials. Finally, we expect to continue the work on summarizing long texts, trying to infer the memory representations that readers acquire during comprehension.